CN115019209A

CN115019209A - Method and system for detecting state of electric power tower based on deep learning

Info

Publication number: CN115019209A
Application number: CN202210697339.6A
Authority: CN
Inventors: 林旭; 李密; 陈旭; 陈佳期; 唐光铁; 曾远强; 卢雨畋; 周小报
Original assignee: Fujian Strait Zhihui Technology Co ltd
Current assignee: Fujian Strait Zhihui Technology Co ltd
Priority date: 2022-06-20
Filing date: 2022-06-20
Publication date: 2022-09-06

Abstract

The invention provides a method for detecting the state of an electric power tower based on deep learning, which specifically comprises the following steps: acquiring data for constructing a deep learning model and dividing a data set; constructing a data module and loading data, and further constructing a loss function of the deep learning model; further constructing a target model for deep learning training, and training and verifying the target model; selecting and obtaining a required target model, and importing the target model into a target power tower to be detected to detect the state of the target model; and outputting a target detection result according to a preset threshold value. Through the combination of deep learning target detection and the detection of the state of the power tower, the generated model is small, and the model is favorable for deployment and transplantation to the edge equipment for result calculation, so that the working efficiency of inspection personnel is improved, and the cost is reduced.

Description

Method and system for detecting state of electric power tower based on deep learning

Technical Field

The invention belongs to the technical field of industrial vision and energy power, and particularly relates to a method and a system for detecting the state of a power tower based on deep learning.

Background

In recent years, irresistible natural disasters such as earthquakes, debris flows and typhoons become objective factors which seriously affect the work and life of people, and especially, the facilities with high gravity centers, such as infrastructure and infrastructure facilities, for example, transmission towers and communication tower lighthouses, are easily affected by external forces of the natural disasters such as the earthquake and typhoons, so that the facilities cannot work normally, and the work and the production of the line and the area are affected.

However, it is known that the cost for inspecting and maintaining the power line of the national power grid is very high every year, and meanwhile, many regions still work in a traditional manual inspection mode, which not only wastes a lot of manpower and material resources, but also brings certain danger to the safety of inspection workers. Therefore, the tower abnormal state inspection of the power line is identified and detected based on the unmanned aerial vehicle inspection, and a solution scheme for detecting and identifying in combination with deep learning is provided.

In view of this, it is very significant to provide a method and a system for detecting the state of a power tower based on deep learning.

Disclosure of Invention

The invention provides a method and a system for detecting the state of an electric power tower based on deep learning, which aim to solve the problems that the existing national power grid is high in cost for routing inspection and maintenance of the electric power line, time and labor are wasted by adopting the traditional manual routing inspection and the like, and solve the technical defect problems in the prior art.

In a first aspect, the invention provides a method for detecting a state of an electric power tower based on deep learning, which specifically comprises the following steps:

s1, obtaining deep learning model data for construction and dividing a data set;

s2, constructing a data module and loading data, and further constructing a loss function of the deep learning model;

s3, further constructing a target model for deep learning training, training and verifying the target model;

s4, selecting and obtaining a required target model, and importing the target model to be detected to a power tower to detect the state of the target model; and

and S5, outputting the target detection result according to the preset threshold value.

Through the combination of deep learning target detection and the detection of the state of the power tower, the generated model is small, and the model is favorable for deployment and transplantation to the edge equipment for result calculation, so that the working efficiency of inspection personnel is improved, and the cost is reduced.

Preferably, S1 further includes:

s11, arranging the unmanned aerial vehicle inspection video acquired by aerial photography, and performing frame taking and sampling according to 30 frames/piece and 60 frames/piece to serve as initial data;

s12, importing initial data, manually drawing a frame through labeling software LabelImg to label the position information of the power tower pole, labeling information of target detection, and representing a labeled text file according to a format of < object-class-ID > < X center > < Y center > < frame width > < frame height >;

and S13, converting the obtained data set in the VOC format into a TXT format, and further dividing the data set into a training set, a verification set and a test set according to the ratio of 8:1:1 to obtain the data set required by the final training.

Further preferably, the deep learning model is constructed based on a YOLOv5 target detection model.

Further preferably, in S2, the loss function of the deep learning model is constructed using the target localization loss box _ loss, the classification loss cls _ loss, and the confidence loss obj _ loss.

Further preferably, in the case of binary classification, the calculation of the loss function is specifically as follows:

J＝-[y

where y represents a swatch label, a positive swatch label is 1, and a negative swatch label is 0.

Preferably, in the case of multi-classification, the calculation of the loss function is specifically as follows:

wherein K is the number of categories and y is the label; i.e. if the category is i, then y _i 1 otherwise y _i 0 is not zero; p is the output of the neural network, i.e., the probability that the class is i.

Preferably, S3 further includes:

s31, setting a pre-training model path, reading the path by a data set, further setting the number of training iterations and the size of an incoming picture, setting other parameter options as defaults, and starting to train the target model;

s32, visualizing the training process and checking the change curve of the indexes related to the model;

s33, after the training is finished, comparing, checking and storing the best target model and outputting the best target model to the designated path;

and S34, performing predictive reasoning on the test sample on the trained target model to verify the target model.

In a second aspect, the present invention further provides a system for detecting a state of an electric power tower based on deep learning, including:

a data acquisition module: configuring and acquiring a data set required by a deep learning model;

a loss function construction module: configuring a loss function of the framework deep learning model;

a calculation module: configuring to calculate a loss function;

a model construction module: and configuring and constructing a target model for deep learning training.

In a third aspect, an embodiment of the present invention provides an electronic device, including: one or more processors; storage means for storing one or more programs which, when executed by one or more processors, cause the one or more processors to carry out a method as described in any one of the implementations of the first aspect.

In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the method described in any implementation manner of the first aspect.

Compared with the prior art, the beneficial results of the invention are as follows:

(1) through the combination of deep learning target detection and the detection of the state of the power tower, the generated model is small, and the model is favorable for deployment and transplantation to the edge equipment for result calculation, so that the working efficiency of inspection personnel is improved, and the cost is reduced.

(2) Through the single-stage target detection model, the speed of the predictive reasoning is effectively improved.

Drawings

The accompanying drawings are included to provide a further understanding of the embodiments and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments and together with the description serve to explain the principles of the invention. Other embodiments and many of the intended advantages of embodiments will be readily appreciated as they become better understood by reference to the following detailed description. The elements of the drawings are not necessarily to scale relative to each other. Like reference numerals designate corresponding similar parts.

FIG. 1 is an exemplary device architecture diagram in which an embodiment of the present invention may be employed;

fig. 2 is a schematic flow chart of a method for detecting a state of an electric power tower based on deep learning according to an embodiment of the present invention;

fig. 3 is a schematic overall flow chart of a method for detecting a state of an electric power tower based on deep learning according to an embodiment of the present invention;

4-1, 4-2, 4-3, and 4-4 are schematic diagrams of a process of collecting a video sorting set, intercepting key frame valid data and storing the key frame valid data in a data set, labeling the data set according to a label, and training in the method for detecting the state of the power tower based on deep learning according to the embodiment of the present invention, respectively;

fig. 5 is a schematic diagram of an architecture of a YOLOv5 target detection model in the method for detecting the state of a power tower based on deep learning according to the embodiment of the present invention;

6-1 to 6-10 are diagrams respectively illustrating a visualization prediction process, labels _ correlogram, the representation of a training set and a verification set on a model, P _ curve, PR _ curve, R _ curve, a positive sample prediction result, a positive sample prediction set sampling result preview, a negative sample prediction result, and a negative sample prediction set sampling result preview in the method for detecting the state of the power tower based on deep learning according to the embodiment of the present invention;

FIG. 7 is a schematic flow chart of a system for detecting the state of a power tower based on deep learning according to an embodiment of the present invention;

FIG. 8 is a schematic diagram of a computer apparatus suitable for use with an electronic device to implement an embodiment of the invention.

Detailed Description

In the following detailed description, reference is made to the accompanying drawings, which form a part hereof, and in which is shown by way of illustration specific embodiments in which the invention may be practiced. In this regard, directional terminology, such as "top," "bottom," "left," "right," "up," "down," etc., is used with reference to the orientation of the figures being described. Because components of embodiments can be positioned in a number of different orientations, the directional terminology is used for purposes of illustration and is in no way limiting. It is to be understood that other embodiments may be utilized and logical changes may be made without departing from the scope of the present invention. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined by the appended claims.

It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Fig. 1 illustrates an exemplary system architecture 100 of a method for processing information or an apparatus for processing information to which embodiments of the present invention may be applied.

As shown in fig. 1, the system architecture 100 may include

terminal devices

101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the

terminal devices

101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

The user may use the

terminal devices

101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The

terminal devices

101, 102, 103 may have various communication client applications installed thereon, such as a web browser application, a shopping application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like.

The

terminal devices

101, 102, 103 may be various electronic devices having communication functions, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

The server 105 may be a server that provides various services, such as a background information processing server that processes check request information transmitted by the

terminal apparatuses

101, 102, 103. The background information processing server may analyze and perform other processing on the received verification request information, and obtain a processing result (e.g., verification success information used to represent that the verification request is a legal request).

It should be noted that the method for processing information provided by the embodiment of the present invention is generally executed by the server 105, and accordingly, the apparatus for processing information is generally disposed in the server 105. In addition, the method for sending information provided by the embodiment of the present invention is generally executed by the

terminal equipment

101, 102, 103, and accordingly, the apparatus for sending information is generally disposed in the

terminal equipment

101, 102, 103.

The server may be hardware or software. When the server is hardware, it may be implemented as a distributed server cluster formed by multiple servers, or may be implemented as a single server. When the server is software, it may be implemented as a plurality of software or software modules (for example, to provide distributed services), or may be implemented as a single software or a plurality of software modules, and is not limited in particular herein.

Fig. 2 shows that an embodiment of the present invention discloses a method for detecting a state of a power tower based on deep learning, and as shown in fig. 2 and fig. 3, the method specifically includes the following steps:

further, S1 specifically includes:

The data set for the deep learning model is divided into several categories according to the executed tasks, namely target detection, image segmentation, image classification and instance segmentation, in the embodiment, the data set with the file in the label format is used for image segmentation, and the data set obtained after labeling is divided into a training set, a verification set and a test set according to the proportion of 8:1: 1.

Specifically, the data set in the VOC format used in this embodiment is mainly to frame and sample the unmanned aerial vehicle polling video acquired by aerial photography, as initial data, label the information of target detection through the label software label img, that is, label the position information of the target object with a manual frame, convert the data set in the VOC format into a txt format dedicated for the YOLO model, and finally divide the proportion according to the data set, thereby obtaining the data set required for final training.

There are usually multiple data sets used to build the final model, and at different stages of building the model there are usually three data sets: training set, validation set and test set.

First, the model is fitted on a training set (training dataset). For supervised learning, the training set is a set consisting of samples used to fit parameters (e.g., weights of links between neurons in an artificial neural network). In practice, a training set is typically a data pair consisting of an input vector (scalar) and an output vector (scalar). Where the output vector (scalar) is called a target or tag. During the training process, the current model predicts each sample in the training set and compares the prediction result with the target. Based on the comparison, the learning algorithm updates the parameters of the model. The process of model fitting may include both feature selection and parameter estimation.

Next, the fitted model is predicted on the second dataset, validation dataset. When adjusting for hyper-parameters of the model (e.g., the number of neurons in the hidden layer in the neural network), the validation set provides an unbiased evaluation of the model fitted on the training set. The validation set can be used for early stopping in regularization: when the validation set error rises, which is a signal that is over-fitted on the training set, the training is stopped. However, in practice, this sometimes fails because the validation set errors fluctuate during the training process. Thus, one has devised rules for determining a signal that is better overfit.

Finally, a test dataset (test dataset) can be used to provide an unbiased evaluation of the final model. If the test set is never used in the training process (e.g., not used in cross-validation), it is also referred to as a reserved set.

Referring to fig. 4-1, 4-2, 4-3, and 4-4, in the present embodiment, the specific processing flow for the data set is as follows:

1. arranging aerial photo inspection videos of the unmanned aerial vehicle;

2. storing the collected and sorted videos into a data set according to 30 frames/sheet and 60 frames/sheet;

3. importing the sorted pictures into labeling software labelImg for data cleaning and data labeling;

4. and converting the marked data file into a yolo format, and training.

And S2, constructing a data module and loading data, and further constructing a loss function of the deep learning model.

Referring to fig. 5, in S2, a loss function of the deep learning model is constructed using the target localization loss box _ loss, the classification loss cls _ loss, and the confidence loss obj _ loss.

The loss function is a function of the distance between the current output and the expected output of the calculation algorithm. It is a method of evaluating how an algorithm models data, which can be divided into two groups, one for classification (discrete values, 0,1, 2.) and the other for regression (continuous values).

In this case it is intended for multi-class classification, where the target values are in the

set

0,1, 3. Mathematically, it is the preferred loss function under the framework of maximum likelihood inference. What is evaluated first is a loss function that changes only if you have sufficient reason.

Cross entropy is the default penalty function for multi-class classification problems. The cross entropy will calculate a score that summarizes the average difference between the actual probability distribution and the predicted probability distribution for all classes in the problem. The fraction is minimized with a perfect cross entropy value of 0.

Specifically, in the present embodiment, the localization loss and the classification loss are calculated by combining the cross-entropy loss function and the logarithmic loss. Cross entropy is a measurement method in the field of information theory, which is based on entropy and usually calculates the difference between two probability distributions.

Cross entropy loss functions are often used in classification problems, particularly neural network classification problems. The cross entropy is used to describe the distance of two distributions, and the goal of neural network training is to approximate g (x) to p (x).

And (3) solving a loss function relative to the sigmoid, on the gradient calculation level, the partial derivative of the cross entropy to the parameters does not contain the derivation to the sigmoid function, and other partial derivatives such as Mean Square Error (MSE) and the like contain the sigmoid function. The gradient is almost zero when the Sigmoid value is small or large, which makes the gradient descent algorithm unable to make effective progress, and the cross entropy avoids this problem.

In order to make up for the defect that the derivative form of the sigmoid-type function is easy to saturate (slow gradient update), Softmax can be introduced as a prediction result to calculate the cross entropy loss. Since cross-entropy involves computing the probability of each class, in neural networks, cross-entropy is closely related to the softmax function.

In the case of binary classification, the final predicted outcome of the model is only 2 classes, with the probabilities we predict being p and 1-p for each class. In this case (Binary Cross Entropy) Binary Cross entry, i.e. the calculation of the loss function, is specifically as follows:

J＝-[y

wherein y represents a sample label, a positive sample label is 1, and a negative sample label is 0;

p represents the probability of prediction as a positive sample.

In the present embodiment, multi-classification is concerned, and in the case of multi-classification, the calculation of the loss function is specifically as follows:

wherein K is the number of categories and y is the label; i.e. if the category is i, then y _i 1 otherwise y _i 0 is not zero; p is the output of the neural network, i.e., the probability that the class is i. This output value is passed through softmax is calculated.

Wherein the input terminal represents an input picture. The size of the input image of the network is 608 x 608, and this stage usually includes an image preprocessing stage, i.e. scaling the input image to the input size of the network, and performing normalization operations. In the network training stage, YOLOv5 uses the Mosaic data to enhance the training speed of the operation lifting model and the accuracy of the network; and provides a self-adaptive anchor frame calculation and self-adaptive picture scaling method.

The reference network is usually a network of classifiers with excellent performance, and the module is used for extracting some universal feature representations. The YOLOv5 uses not only the CSPDarknet53 structure but also the Focus structure as a reference network;

the hack network is usually located in the middle of the reference network and the head network, and the diversity and robustness of the features can be further improved by using the hack network. Although YOLOv5 also uses the SPP module, the FPN + PAN module, the implementation details are somewhat different;

the Head output end is used for finishing the output of the target detection result. For different detection algorithms, the number of branches at the output end is different, and the detection algorithm generally comprises a classification branch and a regression branch. Yolov4 replaces the Smooth L1 Loss function with GIOU _ Loss, thereby further improving the detection accuracy of the algorithm.

Referring to fig. 5, specifically, the deep learning model in the present embodiment is constructed based on the YOLOv5 target detection model. In other embodiments, the deep learning model may also be constructed based on ssd (single Shot detector), detr (detection triansformer). Like the loss function, the YOLOv5 objective function is also composed of three parts:

box _ loss — bounding box regression loss (mean square error);

obj _ loss — confidence that an object exists is the loss of objectionality (binary cross entropy);

cls _ loss-classification loss (cross entropy).

The objective function is three loss functions which are minimized, namely, the final purpose of training the constructed network model is to train a model parameter which enables the three loss functions to be relatively small under the conditions of limited time and data amount and computational resources, so that the inference prediction of other data can be carried out on the basis of the model parameter.

the training and verifying part is mainly used for carrying out corresponding setting according to the hyper-parameter configuration of the network model.

In the training process, several of the important parameters are involved:

data is a data set reading path, epochs is the training iteration times, ingsz is the size of an incoming picture, and other parameters are default; since the data set is divided into the training set, the verification set and the test set when being divided, it is equivalent to verifying the verification set at the same time of the training process.

Referring to fig. 6-1 to 6-10, S3 further includes:

S4, selecting and obtaining a required target model, and importing the target model to be detected into a target power tower to detect the state of the target model; and

Specifically, in this embodiment, the preset threshold is set to 0.45, and the function of the threshold is mainly to limit the magnitude of the confidence, that is, the threshold will filter out the redundant detection results that do not satisfy the value condition > 0.45.

According to the technical scheme, the deep learning target detection is combined with the detection of the state of the power tower, the generated model is small, and the deployment and the transplantation to the edge equipment for result calculation are facilitated, so that the working efficiency of inspection personnel is improved, and the cost is reduced; the speed of prediction and inference is effectively improved through a single-stage target detection model.

In a second aspect, the present invention further provides a system for detecting a state of a power tower based on deep learning, with reference to fig. 7, including:

the data acquisition module 71: configuring and acquiring a data set required by a deep learning model;

loss function building block 72: configuring a loss function of the framework deep learning model;

the calculation module 73: configuring to calculate a loss function;

model building module 74: and configuring and constructing a target model for deep learning training.

Referring now to FIG. 8, a block diagram of a computer apparatus 600 suitable for use with an electronic device (e.g., the server or terminal device shown in FIG. 1) to implement an embodiment of the invention is shown. The electronic device shown in fig. 8 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.

As shown in fig. 8, the computer apparatus 600 includes a Central Processing Unit (CPU)601 and a Graphics Processing Unit (GPU)602, which can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)603 or a program loaded from a storage section 609 into a Random Access Memory (RAM) 606. In the RAM 604, various programs and data necessary for the operation of the apparatus 600 are also stored. The CPU 601, GPU602, ROM 603, and RAM 604 are connected to each other via a bus 605. An input/output (I/O) interface 606 is also connected to bus 605.

The following components are connected to the I/O interface 606: an input portion 607 including a keyboard, a mouse, and the like; an output section 608 including a display such as a Liquid Crystal Display (LCD) and a speaker; a storage section 609 including a hard disk and the like; and a communication section 610 including a network interface card such as a LAN card, a modem, or the like. The communication section 610 performs communication processing via a network such as the internet. The driver 611 may also be connected to the I/O interface 606 as needed. A removable medium 612 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 611 as necessary, so that a computer program read out therefrom is mounted into the storage section 609 as necessary.

In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such embodiments, the computer program may be downloaded and installed from a network via the communication section 610, and/or installed from the removable media 612. The computer programs, when executed by a Central Processing Unit (CPU)601 and a Graphics Processor (GPU)602, perform the above-described functions defined in the method of the present invention.

It should be noted that the computer readable medium of the present invention can be a computer readable signal medium or a computer readable medium or any combination of the two. The computer readable medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor device, apparatus, or any combination of the foregoing. More specific examples of the computer readable medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution apparatus, device, or apparatus. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution apparatus, device, or apparatus. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based devices that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be provided in a processor.

As another aspect, the present invention also provides a computer-readable medium, which may be contained in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to perform the method described in the first aspect of the invention.

The foregoing description is only exemplary of the preferred embodiments of the invention and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention according to the present invention is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is possible without departing from the scope of the invention as defined by the appended claims. For example, the above features and (but not limited to) features having similar functions disclosed in the present invention are mutually replaced to form the technical solution.

Claims

1. A method for detecting the state of an electric power tower based on deep learning is characterized by comprising the following steps:

2. The method for detecting the state of the power tower based on the deep learning of claim 1, wherein S1 further includes:

3. The method for detecting the state of the power tower based on the deep learning as claimed in claim 2, wherein the deep learning model is constructed based on a YOLOv5 target detection model.

4. The method for detecting the state of the power tower based on the deep learning of claim 3, wherein in S2, a loss function of the deep learning model is constructed by using a target location loss box _ loss, a classification loss cls _ loss and a confidence loss obj _ loss.

5. The method for detecting the state of the power tower based on the deep learning as claimed in claim 4, wherein the loss function is calculated specifically as follows under the condition of two classifications:

J＝-[y

6. The method for detecting the state of the power tower based on the deep learning as claimed in claim 4, wherein in the case of multiple classifications, the loss function is calculated as follows:

wherein K is the number of types, and y is a label; i.e. if the category is i, then y _i 1 otherwise y _i 0 is not zero; p is the output of the neural network, i.e., the probability that the class is i.

7. The method for detecting the state of the power tower based on the deep learning of claim 1, wherein S3 further includes:

s32, visualizing the training process and checking the change curve of the index related to the model;

8. The utility model provides a system for electric power tower state detection based on deep learning which characterized in that includes:

a calculation module: configuring to calculate a loss function;

9. An electronic device, comprising:

one or more processors;

storage means for storing one or more programs;

when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-7.

10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-7.